the effect of work experience, training and auditor's

The Effect of Work Experience, Training and Auditor's Expertise on the Detection of

Financial Statement Fraud

Sekolah Tinggi Ilmu Ekonomi Indonesia - 2020 1

THE EFFECT OF WORK EXPERIENCE, TRAINING

AND AUDITOR'S EXPERTISE ON THE DETECTION

OF FINANCIAL STATEMENT FRAUD

1Dedeh Novianingsih, 2Kunarto

Department of Accounting

Sekolah Tinggi Ilmu Ekonomi Indonesia

Jakarta, Indonesia

[email protected]; [email protected]

Abstract - This study aims to examine whether the effect of

work experience, training and expertise of auditors on the

detection of financial statement fraud.

The sample method used was purposive sampling and the

analysis model used was multiple linear analysis with SPSS

26. The data used in this study were primary data collected

through a questionnaire. The questionnaires that are

collected and can be processed are a number of 100

questionnaires from 125 questionnaires distributed.

Respondents of this study are from the Financial and

Development Supervisory Agency (BPKP) and Public

Accounting Firms in the Jakarta area.

The results of this study are that the work experience of

auditors has a significant effect on the detection of fraud in

financial statements, auditor training has a significant effect

on the detection of fraud in financial statements, and auditor

expertise has a significant effect on the detection of fraud in

financial statements. Meanwhile, simultaneously the results

of this study indicate that the effect of work experience,

training and auditor expertise has a significant effect on the

detection of financial statement fraud.

Keywords: Work Experience, Training, Expertise, Fraud

Detection

Abstrak - Penelitian ini bertujuan untuk menguji apakah

Pengaruh Pengalaman Kerja, Pelatihan dan Keahlian

Auditor Terhadap Pendeteksian Kecurangan Laporan

Keuangan.

Metode sampel yang digunakan adalah purposive sampling

dan model analisis yang digunakan adalah analisis linier

berganda dengan SPSS 26. Data yang digunakan dalam

penelitian ini adalah data primer yang dikumpulkan melalui

kuesioner. Kuesioner yang terkumpul dan dapat diolah

adalah sejumlah 100 kuesioner dari 125 kuesioner yang

disebarkan. Responden dari penelitian ini adalah berasal

dari Badan Pengawas Keuangan dan Pembanguan (BPKP)

dan Kantor Akuntan Publik yang ada di wilayah Jakarta.

mailto:dedehnovianingsih7@gmail.

mailto:[email protected]

Dedeh Novianingsih1 , Kunarto2


Hasil dari penelitian ini adalah pengalaman kerja auditor

berpengaruh signifikan terhadap pendeteksian kecurangan

laporan keuangan, pelatihan auditor berpengaruh signifikan

terhadap pendeteksian kecurangan laporan keuangan, dan

keahlian auditor berpengaruh signifikan terhadap

pendeteksian kecurangan laporan keuangan. Sedangkan

secara simultan hasil dari penelitian ini menunjukkan

bahwa pengaruh pengalaman kerja, pelatihan dan keahlian

auditor berpengaruh signifikan terhadap pendeteksian

kecurangan laporan keuangan.

Kata Kunci: Pengalaman Kerja, Pelatihan, Keahlian,

Pendeteksian Kecurangan

I. Introduction

Fraud cases have been in the spotlight for all circles in society, especially in cases related

to financial reporting problems involving companies, both large and small companies. The

perpetrators who commit fraud are not only from upper class employees, but also many of them

have reached lower class employees. Of course, this must be one of the things we need to be aware

of and care about for the environment in which we work.

According to The Association of Certified Fraud Examiners (ACFE, 2016), fraud are

actions that are against the law that are carried out on purpose (manipulation or giving false reports

against other parties) by people from inside or outside the organization to obtain personal or group

gain either directly or indirectly harming other parties. The Association of Certified Fraud

Examiners (ACFE) classifies fraud into three forms based on actions, namely asset

misappropriation, fraudulent financial reporting and corruption.

Based on the results of a survey conducted by ACFE Indonesia, the most prevalent fraud in

Indonesia is corruption, amounting to 67%. In contrast to the results of the Report to The Nationss

(2016) issued by ACFE which states that the most types of fraud are found in the form of asset

misappropriation. In the Indonesian fraud survey, the fraud found in the form of asset

misappropriation was 31%. Fraud in the form of financial statements is the third largest type of

fraud at 2%. This difference is caused by different knowledge and experiences of the respondents.

Of the three cases, the fraud case against financial statements was considered the most

damaging thing. This is because in Indonesia, from various kinds of crimes, crimes against

financial statements have not been widely revealed. For example, a recent case, namely fraudulent

financial statements committed by PT. Garuda Indonesia (Persero). The incident began with the

results of the financial statements of PT. Garuda Indonesia for financial year 2018. In the financial

statements, Garuda Indonesia Group recorded a net profit of USD 809.85 thousand or the

equivalent value of rupiah, namely IDR 11.33 billion (assuming an exchange rate of IDR 14,000

per US dollar). This figure is soaring high when compared to 2017 which suffered a loss of USD

216.5 million. This led to a polemic between the two commissioners of Garuda Indonesia, namely

Chairal Tanjung and Dony Oskaria (who are currently no longer in office), considering that

Garuda's 2018 financial statements were not in accordance with the Statement of Financial

Accounting Standards (PSAK). The reason is, Garuda includes the benefits of PT. Mahata Aero

Teknologi which owes debt to the red-plate airline. Mahata itself still has debts related to installing

unpaid wifi (https://economy.okezone.com).

From the case example above, if during the auditing process the auditor finds indications of

fraud, it is the auditor's responsibility to communicate this immediately to the appropriate level of

https://economy.okezone.com/




management, with the aim of informing those who have the primary responsibility for preventing

and detecting fraud about this matter - things that are relevant to their responsibilities (SPAP, 2014:

240.1-14).

Auditors who have work experience with high flying hours carry out audit procedures

related to giving opinions on the audit report Kushasyandita (2012: 3) in Irawan et. al. (2018).

Auditors who have a lot of experience will not only have the ability to find errors or fraud that are

not common in the financial statements but also these auditors can provide a more accurate

explanation of their findings compared to auditors who are still small experience. In addition,

auditors also need certain skills and expertise to improve their performance, especially in detecting

fraud, therefore they require training through advanced professional education courses. The more

auditors carry out training, the more specific the knowledge they have and the easier it will be to

detect fraud and increase their responsibility in detecting fraud. Then the auditor's expertise can be

further developed by attending auditor training and certification where he will learn new things that

will expand his knowledge and abilities. Expertise helps auditors in the engagement process

because the more skilled an auditor is, his ability to detect fraud and minimize procedural errors is

better. Then the auditor's expertise can be further developed by attending auditor training and

certification where he will learn new things that will expand his knowledge and abilities. Expertise

helps auditors in the engagement process because the more skilled an auditor is, his ability to detect

fraud and minimize procedural errors is better. Then the auditor's expertise can be further

developed by attending auditor training and certification where he will learn new things that will

expand his knowledge and abilities. Expertise helps auditors in the engagement process because the

more skilled an auditor is, his ability to detect fraud and minimize procedural errors is better.

II. THEORETICAL FRAMEWORK AND HYPOTHESIS DEVELOPMENT

Auditor Work Experience

Experience is a process of learning and enhancement of potential development and a

process that brings a person to something higher.

According to Mulyadi (2010: 41) the definition of auditor experience is the combined

accumulation of everything obtained through interaction. Audit experience is shown by the

auditor's flying hours in carrying out audit procedures related to providing an opinion on the audit

report, whereas according to Suryani & Helvinda (2015), auditor experience is the ability of the

auditor to learn from past events related to an audit or examination.

Sucipto's research (2007) in Eko (2014) states experience is knowledge or expertise gained

from an event through direct observation or participation in the event. So, an auditor with more

flying hours will be able and accustomed to finding fraud carefully compared to an auditor whose

flight hours are still few.

This is supported based on the results of his research which states that work experience has

a positive effect on the ability of auditors to detect fraud. Meanwhile, according to research by

Friska et al. (2019) stated that the audit experience does not have a positive and significant effect

on the ability of auditors to detect fraud. Likewise, research according to Ida and Dewa (2016)

suggests that the experience of auditors has a positive effect on fraud detection.

H1: Auditor's Work Experience has a positive effect on Fraud Detection

Auditor Training

Training is one of the developments in resources, especially in terms of knowledge, skills,

abilities and attitudes. According to Samsudin (2014: 111) states that training is an attempt to

improve mastery of various work skills in a relatively short (short) time. Meanwhile, auditor

training is an effort to develop human resources in the fields of knowledge, abilities, expertise and



attitudes of Sanjaya (2017).

Auditor training is an effort to develop human resources in the fields of knowledge,

abilities, expertise and attitudes. With the existence of systematic and tiered training according to

the auditor's level, it will make it easier for auditors to complete auditor shortages and emphasize

auditing practices and accounting standards for auditors. Auditors need a variety of certain skills

and expertise in improving their performance, especially in detecting fraud, therefore they require

training through further professional education courses. The more often the auditors attend training,

the more auditors will develop specific knowledge about the audit field,

This research is supported in Wudu (2014) which states that auditor training has a positive

effect on the responsibility of auditors in detecting fraud. This is also supported by Hilmi's research

(2011) which states that auditor training has a positive effect on the responsibility of auditors in

detecting fraud. Meanwhile, research by Friska et al. (2019) stated that audit training had no

positive and significant effect on the ability of auditors to detect fraud. This statement is the same

as research by Linda et. al (2019) stated that fraud audit training does not have a significant effect

on the ability of auditors to detect fraud. So it can be concluded that the higher the level of fraud

audit training, the auditor's ability to detect fraud increases.

H2: Auditor training has a positive effect on Fraud Detection

Auditor's Expertise

According to Sanger et. al. (2016) expertise is a basic competency that must be possessed

by an auditor who can assist in carrying out the audit process. Expertise is an important element

that must be possessed by an independent auditor to work as a professional. Professional character

is the state of perfection in a technique that a person has through years of practice and study which

is useful for developing the technique, and the desire to achieve perfection and excellence over his

peers. So, the true professional must have a clear character and broad experience. Services provided

by clients must be obtained in professional ways which are obtained by learning, training,

experience and refinement of auditing skills.

H3: Auditor's expertise has a positive effect on Fraud Detection

III. RESEARCH METHODS

Population and Sample

Population is a generalization area consisting of objects / subjects that have certain

quantities and characteristics set by researchers to study and then draw conclusions, Sugiyono

(2018: 136). The population in this study is an auditor, both internal auditors who work at the

Jakarta Financial and Development Supervisory Agency (BPKP) and external auditors who work at

the Public Accounting Firm (KAP) in the Jakarta area for at least 1 year.

The sampling method used in this study is a non probability sampling method. Non

probability sampling is a sampling method that is taken based on the availability of elements and

the ease of obtaining them. The reason for using this method is because the population size is

unknown. According to Sujarweni (2015: 155) because the population is unknown, the minimum

sample size is determined by the formula:

n = Z2/4(Moe)2

Where:

n = Number of samples

Z = The level of normal distribution at the significant level of 5% = 1.96

Moe = Margin of Error Max, ythat is the maximum error rate in sampling still

can be tolerated or desired




By using a max margin of error of 10%, the minimum sample size that can be taken is:

n = (1.96)2/4(0.10)2

n = 96.04 or 97

The sampling technique uses one of the non-probability sampling methods, namely

purposive sampling, in which the sample selection is carried out based on the objectives and

intentions of the researcher choosing samples based on certain criteria. The sample criteria in this

study are:

1. An external auditor who works at the Public Accounting Firm (KAP) in the Jakarta area.

2. An internal auditor who works at the DKI Jakarta Province Financial and Development

Supervisory Agency (BPKP).

3. An auditor who has at least 1 year of experience.

Operationalization of Variables


According to Eko's research (2014), experience is knowledge or expertise gained from an

event through direct observation or participation in the event. An experienced auditor who is able

to detect, understand and find the causes of fraud. That way, an auditor with more flying hours will

be able and accustomed to finding fraud carefully compared to auditors with few hours of flying

time. Based on this understanding, the Auditor Work Experience (X1) variable can be measured

from several points, namely:

a. The length of time he worked as an auditor

b. The number of assignments completed in one year

c. Types of companies handled in one year

Auditor Training

Mangkunegara (2014: 41) says training is a short-term educational process that uses a

systematic and organized procedure in which non-managerial employees learn technical knowledge

and skills for limited purposes. Samsudin (2014: 111) states that training is an attempt to improve

mastery of various work skills in a relatively short time (short). There are five factors that cause the

need for training, namely as follows: the quality of the workforce, global competition, rapid and

continuous change, the problem of technology transfer and demographic changes.

In carrying out this training there are several factors that need to be considered and play a

role in carrying out the training, namely: instructors, participants, materials, methods, principles of

learning and training evaluation. In this study, Auditor Training (X2) was measured based on

material, methods, principles of learning and training evaluation.

Auditor's Expertise

According to M. Rifki Ismail (2018) expertise is part of internal attribution whose

existence is largely determined by factors from within the individual including ability and effort.

Individuals who try with all their abilities to improve their skills will have better knowledge so that

they will respond better to the social perceptions around them. Auditors with more expertise will be

better at understanding the signs of fraud (red flags) that occur around them. So, Auditor Expertise

(X3) is measured based on two aspects, namely structural aspects and attitude aspects

(Martondang, 2010).

Financial Statement Fraud Detection

According to Mokoagouw et al. (2018), fraud detection is related to an auditor's knowledge

of fraud and an auditor's ability to detect fraud. So, in this study, fraud detection (Y) is measured

based on an auditor's knowledge of fraud and an auditor's ability to detect fraud.



Data analysis method

The data analysis technique in this study uses quantitative analysis to better understand the

identification of fraud in audits. In addition, the quantitative analysis method uses a questionnaire

which contains open-ended questions so that the resulting data is richer, humane, sharp, and often

more insightful. This research uses multiple linear regression analysis method through SPSS

version 26 program. Multiple linear regression analysis can be seen in the following equation:

𝑌 = 𝑎 + 𝑏1𝑥1 + 𝑏2𝑥2 + 𝑏3𝑥3 + 𝑒 Information:

𝑌 = The dependent variable is Financial Statement Fraud Detection

𝑎 = Constant

𝑏1 − 𝑏3 = 1st to 3rd independent variable regression coefficient

𝑥1 = Auditor Work Experience

𝑥2 = Auditor Training

𝑥3 = Auditor's expertise

𝑒 = Standard Error

IV. RESULTS AND DISCUSSION

DThe data used in this research is a questionnaire that has been filled in by auditors who

work at BPKP and KAP Daerah DKI Jakarta Province. Respondents who have participated and

filled out a questionnaire as many as 100 respondents and all of them can be processed.

Table 1: List of Research Sample Distribution

No. Name of Agency / Company / KAP Territory Number of

Questionnaires

1 Finance and Development Supervisory

Agency (BPKP) East Jakarta 47

2 KAP Drs. Bambang Sudaryono & Partners East Jakarta 3

3 KAP Heru, Saleh, Marzuki & Partners East Jakarta 15

4 KAP Shohibul, Kaslani, Komarianto &

Santosa East Jakarta 5

5 RI Financial Supervisory Agency Central Jakarta 16

6 Financial Fervices Authority Central Jakarta 1

7 KAP Purwantoro, Sungkoro & Surja (EY

Indonesia) South Jakarta 9

8 KAP Sriyadi Elly Sugeng & Partners South Jakarta 1

9 KAP HLB Hadori Sugiarto Adi & Partners South Jakarta 1

10 KAP Rudi North Jakarta 2

Source: Primary data processed, 2020

Research Result

The validity test is used to measure the validity of a questionnaire (Ghozali, 2016).

Significant test is done by comparing the value of rcount (Corrected Item-Total Correlation value at

the Cronbach alpha output) with the rtable value for degree of freedom (df) = n - 2. With the number

of samples (n) is 100 and a significant level of 0.05. Then the r table in this study is as follows:

Degree of Freedom (df) = n - 2

= 100 - 2

= 98 seen from the distribution of rtabel values

rtable = 0.165




If rcount is greater than rtable and positively correlated, the question item is said to be valid.

The results of the validity test in this study are as follows:

Table 2: Results of the Validity Test for Fraud Detection Variable (Y)

Question Points Item-Total

Correlation R Table Information

Y1 0.583 0.165 Valid

Y2 0.658 0.165 Valid

Y3 0.729 0.165 Valid

Y4 0.600 0.165 Valid

Y5 0.612 0.165 Valid

Y6 0.375 0.165 Valid

Y7 0.506 0.165 Valid


Table 3: Results of the Validity Test of the Auditor Work Experience Variable (X1)



X1.1 , 583 0.165 Valid

X1.2 , 641 0.165 Valid

X1.3 , 523 0.165 Valid


Table 4: Results of the Validity Test of Auditor Training Variables (X2)



X2.1 , 652 0.165 Valid

X2.2 , 753 0.165 Valid

X2.3 , 652 0.165 Valid

X2.4 , 596 0.165 Valid

X2.5 , 651 0.165 Valid


Table 5: Results of the Validity Test of the Auditor Expertise Variable (X3)



X3.1 0.260 0.165 Valid

X3.2 0.455 0.165 Valid

X3.3 0.383 0.165 Valid

X3.4 0.460 0.165 Valid

X3.5 0.468 0.165 Valid


Based on the tables above, the results of the validity test show that all question items in the

questionnaire have item-total correlation> 0.165, so the question is declared valid.

The reliability test was carried out on the question items which were declared valid. This

test is used to measure a questionnaire which is an indicator of a variable or construct. A variable is

said to be reliable if it has Cronbach's Alpha> 0.60 (Ghozali, 2016). The results of the calculation



of the reliability coefficient value for this research instrument are as follows:

Table 6: Reliability Test Results

Variable Cronbach's Alpha N of Item Information

Fraud Detection 0.819 7 Reliable

Auditor Work Experience 0.748 3 Reliable

Auditor Training 0.846 5 Reliable

Auditor's Expertise 0.644 5 Reliable


Based on Table 6, the results obtained from the reliability test in this study, because the

reliability coefficient value is> 0.60, it can be concluded that the research instrument is declared

reliable.

The normality test is carried out to test whether in a regression model, confounding

variables or their residuals are normally distributed or not. This test can be seen from graph

analysis and statistical analysis.

Graph analysis can be seen from the SPSS output results from the histogram and normal

PP Plot charts. The image below is a histogram graph to test the normality in this study.

Figure 1. Hsitogram graph

Source: SPSS output (data processed, 2020)

From the Histogram Graph in Figure 1 above, it can be seen that the histogram graph

follows a bell-shaped line pattern. This illustrates in accordance with the basis for decision making

where the Regrei model meets the normality assumption. So it can be concluded that the data has a

normal distribution line and can be tested further. In addition, it can also be seen from the normal

PP Plot graph for other normality tests. The following is a normal PP Plot graphic image in this

study.




Figure 2. Normal PP Plot Graph


Based on the PP Plot Normal Graph in Figure 2 above, it can be seen that the points spread

and follow the direction around the diagonal line, and there are points that spread some distance

from the diagonal line. This happens because of the different elements of subjectivity.

The next normality test is statistical analysis which is used to provide further explanation if

there is an error in interpretation through graphic analysis. This test can be seen from the

Kolmogorov-Smirnov non-parametric test with the following test criteria:

1. If the significant value (Asymp.Sig)> 0.05, the residual data is normally distributed.

2. If the significant value (Asymp.Sig) <0.05, the residual data is not normally distributed.

Table 7: Kolmogorov-Smirnov Non Parametic Test Results One-Sample Kolmogorov-Smirnov Test

Unstandardized

Residual

N 100

Normal Parametersa, b Mean , 0000000

Std. Deviation 2,30016056

Most Extreme Differences Absolute , 071

Positive , 056

Negative -, 071

Statistical Test , 071

Asymp. Sig. (2-tailed) , 200c, d

a. Test distribution is Normal. b. Calculated from data. c. Lilliefors Significance Correction. d. This is a lower bound of the true significance.


Based on Table 7, the Kolmogorov-Smirnov non-parametric test shows a statistical test of

0.071 and significant 0.200, this means it fulfills the normality test because it has a significant



value> 0.05 and is normally distributed. So it can be concluded that this regression model can

fulfill the assumption of normality.

In research (Dewanoko, 2018) the regression model is said to fulfill the assumption of

normality if the data spreads around the diagonal line and follows the direction of the diagonal line

or the histogram graph (Ghozali, 2013).

The multicollinearity test is needed to determine whether or not there are independent

variables that have similarities between the independent variables in a model. The similarity

between the independent variables will result in a very strong correlation. In addition, this test is

also to avoid habits in the decision-making process regarding the effect of the partial test of each

independent variable on the dependent variable. If the resulting VIF is between 1-10 then there is

no multicollinearity.

Table 8: Multicollinearity Test Results

Coefficientsa

Collinearity Statistics

Model Tolerance VIF

1 (Constant)


, 722 1,385

Auditor Training , 619 1,616 Auditor's Expertise , 680 1,471

a. Dependent Variable: L / K Fraud Detection


Based on Table 8 above, it can be concluded that the VIF number on the independent

variable for VIF auditor work experience is 1.385, VIF auditor training is 1.616 and VIF auditor

expertise is 1.471, so the regression model proposed does not occur multicollinearity.

The autocorrelation test in a model aims to determine whether there is a correlation

between confounding variables in a certain period and the previous variable. For time series data,

autocorrelation often occurs. But for the sample data, crossection rarely occurs because the

confounding variables are different from one another. Detect autocorrelation by using the Durbin

Watson value with criteria if:

1. DW number below -2 means there is positive autocorrelation

2. A DW number between -2 and +2 means that there is no autocorrelation

3. DW numbers above +2 means that there is negative autocorrelation

Table 9: Autocorrelation Test Results Model Summary b

Model R R Square Adjusted R

Square Std. Error of the

Estimate Durbin-Watson

1 , 663a , 439 , 422 2,336 1,601

a. Predictors: (Constant), Auditor Expertise, Auditor Work Experience, Auditor Training

b. Dependent Variable: L / K Fraud Detection


Based on Table 9 above, the results of the analysis show that the DW number is +1,601.

This means that the research model does not have autocorrelation problems.

Heteroscedasticity is testing the difference in residual variance from one observation period

to another. How to predict the presence or absence of heteroscedasticity in a model can be seen




with the Scatterplot image pattern, a regression that does not occur heteroscedasticity if the data

points spread above and below or around the 0 on the Y-axis, the data points do not collect only

above or below alone, the distribution of data points should not form a wavy pattern, widened then

narrowed and widened again, the distribution of data points is not patterned.

Figure 3.Heteroscedasticity Test Results


Based on the Scatterplot Graph in Figure 3, it can be seen that the data points spread up and

below 0 on the Y axis, and the distribution of the points is neither patterned nor spread randomly.

Then it can be assumed that there is no heteroscedasticity in the regression model so that the

regression model is suitable for use.

Data analysis used multiple linear regression analysis to test the independent variable,

namely the effect of work experience, training and auditor expertise on the dependent variable,

namely the detection of financial statement fraud. The multiple linear regression equation in this

study can be seen from Table 9.

Table 10: Results of Multiple Linear Regression Analysis Coefficientsa

Model

Unstandardized Coefficients

Standardized Coefficients

T Sig. B Std. Error Beta

1 (Constant) 8,188 2,729 3,000 , 003


, 250 , 124 , 181 2,008 , 047

Auditor Training , 292 , 124 , 229 2,355 , 021

Auditor's Expertise , 611 , 143 , 395 4,262 , 000



From the table above, the multiple linear regression equation model in this study can be

formulated as follows:

𝑃𝐾𝐿𝐾 = 8,188 + 0,250𝑃𝐾𝐴 + 0,292𝑃𝐴 + 0,611𝐾𝐴 + 𝑒



Information:

PKLK = Financial Statement Fraud Detection

PKA = Auditor Work Experience

PA = Auditor Training

KA = Auditor's expertise

From this equation, it can be seen the influence of the variables of Work Experience,

Auditor Training and Expertise on the detection of fraudulent financial statements. The meaning of

the above equation can be explained as follows:

a. In the multiple linear regression equation above, it can be concluded that the constant value

is 8,188 which means that if there are no changes in the independent variables (PKA, PA,

KA) that affect the PKLK value is 8,188.

b. The auditor work experience variable has a regression value of 0.250. Where the regression

value which is positive indicates that if the auditor's work experience increases by 1%, it

will increase the detection of fraudulent financial statements by 0.250, assuming that other

variables are constant.

c. The auditor training variable has a regression value of 0.292. Where the regression value

which is positive indicates that if the auditor training increases by 1%, it will increase the

detection of fraudulent financial statements by 0.292 with the assumption that other

variables are constant.

d. The auditor expertise variable has a regression value of 0.611. Where the regression value

which is positive indicates that if the auditor's work experience increases by 1%, it will

increase the detection of fraud in financial statements by 0.611 with the assumption that

other variables are constant.

Partial test (t test) is a test of individual partial regression coefficients used to determine

whether the independent variable (X) individually affects the dependent variable (Y). The

following t test results are shown in table form below:

Table 11: Partial Test Results (t test) Coefficientsa

Model

Unstandardized Coefficients

Standardized Coefficients

T Sig. B Std. Error Beta

1 (Constant) 8,188 2,729 3,000 , 003


, 250 , 124 , 181 2,008 , 047

Auditor Training , 292 , 124 , 229 2,355 , 021

Auditor's Expertise , 611 , 143 , 395 4,262 , 000



Based on the table above, the hypothesis test results show that the t-table value obtained

from the t distribution table is sought at the 0.05 level. Because it uses two sides, the actual level is

0.05 / 2 = 0.025. The number of samples in this study was 100 (n = 100), 3 (k = 3) independent

variables, and degree of freedom (df) = nk-1, then the t table can be formulated as follows:

ttable = t (α / 2) (nk-1)

= t (0.05 / 2) (100-3-1)

= t (0.025) (96)

ttable = 1.98498

The Auditor Work Experience variable has a tcount of 2.008, this value is greater than the

ttable value of 1.98498 and a Sig t value of 0.047 which is smaller than 0.05, so H0 is rejected. This




may imply that partially the variable "Auditor Work Experience affects the Detection of Financial

Statement Fraud" is accepted.

The Auditor Training variable has a tcount of 2.355, this value is greater than the ttable value

of 1.98498 and a Sig t value of 0.021 which is smaller than 0.05, then H0rejected. This means that

partially the variable "Auditor Training affects the Detection of Financial Statement Fraud" is

accepted.

The Auditor Expertise variable has a tcount of 4.262, this value is greater than the ttable value

of 1.98498 and a Sig t value of 0.000 which is smaller than 0.05, then H0rejected. This may imply

that partially the variable "Auditor expertise affects the Detection of Financial Statement Fraud" is

accepted.

Simultaneous test (F test) was conducted to determine the independent variables together

on the dependent variable. The results of the F test in this study can be seen in the table below.

Table 12: Simultaneous Test Results (Test F) ANOVAa

Model Sum of Squares Df Mean Square F Sig.

1 Regression 410,007 3 136,669 25,049 , 000b

Residual 523,783 96 5,456

Total 933,790 99

a. Dependent Variable: L / K Fraud Detection b. Predictors: (Constant), Auditor Expertise, Auditor Work Experience, Auditor Training


Based on the above, the results of the regression analysis show that Fcount = 25.049

α = 5%; df1 = k-1; df2 = nk

Where n = 100; k = 4

df1 = 3; df2 = 96

Then Ftable = 2.70

And from the calculation results in Table 12 shows that the value of Fcount is 25.049 (Sig f

= 0.000) and the value of Ftable with the real level (α) = 5% and df1 = 3 and df2 = 96 is 2.70 so it

can be concluded that Fcount> Ftable (25.049> 2.70) and Sig f <0.05 (0.000 <0.05). Thus H4 is

accepted and H0 is rejected, which means that together the Work Experience, Training and Auditor

Skills variables have a significant effect on the Fraud Detection variable.

The coefficient of determination is used to determine the percentage of the influence of the

independent variable on the dependent variable. The coefficient of determination is between zero

and one, if the smaller the adjusted R2 value means the ability of the independent variables to

explain the dependent variable is very limited and vice versa. The following are the results of the

determination coefficient test in this study.

Table 13: Results of the Coefficient of Determination (R2) Model Summary b

Model R R Square Adjusted R

Square Std. Error of the

Estimate

1 , 663a , 439 , 422 2,336

a. Predictors: (Constant), Auditor Expertise, Auditor Work Experience, Auditor Training b. Dependent Variable: L / K Fraud Detection




Based on Table 13, it can be seen that the R2 value is 0.439. This value indicates that

43.9% of the variation in the detection of fraud in financial statements can be explained by

variations in work experience, training and auditor expertise variables.

Discussion

The Effect of Auditor's Work Experience on Fraud Detection of Financial Statements

The test results of the First Hypothesis (H1) show that the Auditor's Work Experience has a

significant effect on the Detection of Financial Statement Fraud. It can be seen from Table 11 that

it is known that the t-count value of 2.008, this value is greater than the t-table value of 1.98498

and the Sig t value of 0.047 which is smaller than 0.05. This shows that the results of this study

support the First Hypothesis (H1) regarding "Auditor's Work Experience on Financial Statement

Fraud Detection".

Work experience that is measured by the length of work, the number of audit tasks and the

type of company handled, the audience will have a lot of experience to get good audit results. From

the data above, it shows that work experience has an influence on fraud detection and it can be said

that some of the auditors who work in the Jakarta area have good experiences. The more experience

the auditor has, the better and he will understand how to deal with problems with the object of the

audit examination, and vice versa.

This is in line with research from Indri et al (2018) which shows that experience has an

influence on fraud detection. Then research from Elly and Vanya (2012) states that partially

experience has an influence on fraud detection by auditors.

Effect of Auditor Training on Fraud Detection of Financial Statements

The test results of the Second Hypothesis (H2) indicate that Auditor Training has a



and the Sig t value of 0.021 which is smaller than 0.05. This indicates that the results of this study

support the Second Hypothesis (H2) regarding "Auditor Training on Financial Statement Fraud

Detection".

Training is measured from training materials, training methods, learning principles and also

evaluation of training, so that auditors who have done a lot of training will find it easier to

understand the types of fraud and carry out correct audit procedures. From this amount of training,

it will increase the auditor's expertise in detecting fraud in financial statements. From the test

results above, it shows that training has an influence on fraud detection and it can be said that some

of the auditors who work in the Jakarta area have done a lot of training. The more auditors attend

training, the easier it will be for an auditor to understand the types of fraud and to follow audit

procedures correctly in detecting fraud.

This is in line with the research of Hilmi (2011) and Wudu (2014) which states that auditor

training has a positive effect on auditor responsibility in detecting fraud.

The Effect of Auditor's Expertise on Fraud Detection of Financial Statements

The test results of the Third Hypothesis (H3) show that the Auditor's Expertise has a



and the Sig t value of 0.000 which is smaller than 0.05. This indicates that the results of this study

support the Third Hypothesis (H3) regarding "Auditor's Expertise on Financial Statement Fraud

Detection".

Expertise is measured based on structural and attitude aspects, so the auditor who has a lot

of expertise will be better for the auditor in understanding the signs of fraud. So that individuals

who always try to improve their skills with all their abilities will have better knowledge so that they

are able to answer the perceptions of those around them will be better. So it can be said that




expertise has an influence on fraud detection and that some of the auditors who work in the Jakarta

area have very good expertise. So it can be concluded that the third hypothesis can be accepted,

namely the Auditor's Expertise in Detecting Fraud in Financial Statements.

This is in line with research from Indri et al (2018) which states that expertise has an

influence on fraud detection. Likewise with research from Fikri (2018) which states that expertise

has a significant effect on the ability of auditors to detect fraud.



The test results of the Fourth Hypothesis (H4) show that the work experience, training and

expertise of auditors have a significant effect on the Detection of Financial Statement Fraud. It can

be seen from Table 12 that it is known that the Fcount value is 25.049 (Sig f = 0.000) and the

Ftable value with the real level (α) = 5% and df1 = 3 and df2 = 96 is 2.70 so it can be concluded

that Fcount> Ftable ( 25.049> 2.70) and Sig f <0.05 (0.000 <0.05). This shows that simultaneously

the results of this study support the "Work Experience, Training and Expertise of Auditors on

Financial Statement Fraud Detection". Because it can be said that in detecting fraud against

financial statements, an auditor must have sufficient work experience,

V. CONCLUSION, SUGGESTION AND LIMITATIONS OF THE RESEARCH

Conclusion

Based on the results of the research that has been done, it can be concluded as follows:

1. The partial test results show that the Auditor's Work Experience has a significant effect on

the Detection of Financial Statement Fraud. Based on these results, it indicates that the

more experience the auditor has, the better and understand how to deal with problems with

the object of the audit examination in detecting fraud on financial statements.

2. The partial test results show that Auditor Training has a significant effect on the Detection

of Financial Statement Fraud. Based on these results, it indicates that the more auditors

attend training, the easier it will be for an auditor to understand the types of fraud and to

follow audit procedures correctly in detecting fraud on financial statements.

3. The partial test results show that the Auditor's Expertise has a significant effect on the

Detection of Financial Statement Fraud. Based on these results, it indicates that the more

an auditor who has sufficient expertise, the faster and more responsive will be in detecting

fraud on financial statements.

4. The simultaneous test results show that the work experience, training and expertise of

auditors have a significant effect on the Detection of Financial Statement Deficiencies.

Based on these results indicate that in detecting fraud on financial statements, an auditor

must have sufficient work experience, training and sufficient expertise so that he can carry

out audit tasks properly and correctly.

Suggestion

Based on the results of the conclusions that have been stated above, the researcher provides

several suggestions as follows:

1. For future researchers, it is recommended that you add other variables that can affect the

detection of fraud in financial statements. So that it can expand knowledge and better

understanding.

2. For agencies or public accounting firms, it is advisable to continue to provide support and

assistance to auditors in carrying out audit tasks.

3. For public accountants or auditors, it is advisable to maintain their professional attitude as

an auditor. So that it can increase expertise and better experience in auditing.



Research Limitations

This research has been attempted by researchers to be carried out properly and according to

scientific procedures, but it still has limitations that can be used as a reference for further research

so as to obtain better results. The following are the limitations of this study:

1. Of the several factors that can affect the detection of fraud, researchers only use the work

experience, training and expertise of auditors so that they cannot be explained more

broadly in this study.

2. This research is limited because it only uses several public accounting firms as

representatives to become research respondents.

3. This study takes a long time to obtain data according to the desired sample size.

4. This research is inadequate because the current condition is precarious with the outbreak

that has occurred so that it is difficult to find a place to study because of the

implementation of WFH and WFO.

REFERENCE

Adnan, Jahari dan Kiswanto. 2017. Determinant of Auditor Ability to Detect Fraud with

Professional Sceptisism as A Mediator Variable. Accounting Analysis Journal, 6 (3), 313-

325.

Afiani, Friska Ayudia. et. al. 2019. Skeptisme Profesional, Pelatihan Audit Kecurangan,

Pengalaman Audit dan Kemampuan Auditor dalam Mendeteksi Kecurangan pada

Inspektorat Kota dan Kabupaten di Jawa Tengah. Prosiding Mahasiswa Seminar Nasional

Unimus, 2, 564-571.

Arens A. Alvin, Randal J. Elder dan Mark S. Beasley. 2015. Auditing dan Jasa Assurance

Pendekatan Terintegrasi. Jilid 1, Edisi 15. Jakarta: Erlangga.

Association of Certified Fraud Examiners. 2017. Survei Fraud Indonesia 2016. Jakarta : ACFE

Indonesia Chapter (https://acfe-indonesia.or.id).

Biksa, Ida A. I. dan I. D. N. Wiratmaja. 2016. Pengaruh Pengalaman, Indepedensi, Skeptisme

Profesional Auditor pada Pendeteksian Kecurangan. E-Jurnal Akuntansi Universitas Udayana, 17

(3), 2384-2415.

E. Kieso, Donald, Jerry J, Weygandt and Teery D. Warfield. 2011. Intermediate Accounting, Edisi

12. Jakarta: Erlangga.

Fakhruddin et. al. 2018. Effect of Expertise Independence and Professional Skepticism About The

Ability of Internal Auditors to Detect Fraud (Examine Emperically on Inspectorate of Bima

Regency and Bima City West Nusa Tenggara Province). International Conference and Call

for Paper, 1185-1208.

Faktor yang Mendorong & Cara Mendeteksi Terjadinya Fraud atau Kecurangan. (2020, Januari

07). www.jurnal.id/id/blog/faktor-fraud/

Fikri, Hasnul. 2019. Pengaruh Independensi, Objektivitas, Pengetahuan dan Pengalaman Kerja

Terhadap Kualitas Audit (Survei pada Inspektorat Kabupaten Bandung). Skripsi.

Institutional Repositories & Scientific Journals. Bandung: Perpustakaan Fakultas Ekonomi

dan Bisnis UNPAS.

Ghozali, Imam. 2013. Aplikasi Analisis Multivariate dengan Program IBM SPSS 21 Update PLS

Regresi. Semarang: Badan Penerbit Universitas Diponegoro.

Ghozali, Imam. 2016. Aplikasi Analisis Multivariate dengan Program IBM SPSS 22. Semarang:

Badan Penerbit Universitas Diponegoro.

Ghozali, Imam. 2018. Aplikasi Analisis Multivariate dengan Program IBM SPSS 25, Edisi 9.

Semarang: Badan Penerbit Universitas Diponegoro.

Hilmi, Fakhri. 2011. Pengaruh Pengalaman, Pelatihan dan Skeptisisme Profesional Auditor

Terhadap Pendeteksian Kecurangan). Skripsi. Jakarta: Universitas Islam Negeri Syarif

https://acfe-indonesia.or.id/

http://www.jurnal.id/id/blog/faktor-fraud/




Hidayatullah.

Indrawati, Linda. et. al. (2019). Pengaruh Skeptisisme Profesional, Independensi Auditor dan

Pelatihan Audit Kecurangan Terhadap Kemampuan Auditor dalam Mendeteksi

Kecurangan. International Journal of Social and Business, 3 (4), 393-402.

Irawan, Kristian Fernando et. al. (2018). Analisis Pengaruh Pengalaman Audit, Beban Kerja,

Skeptisme Profesional, dan Independensi Terhadap Kemampuan Auditor Mendeteksi

Fraud. Jurnal Akuntansi dan Sistem Teknologi Informasi, Vol.14, Edisi Khusus, 146-160.

Irdawanti dan Paulus Uppun. 2018. Pengaruh Keahlian dan Independensi Auditor Dalam

Profesional Judgment Terhadap Pendeteksian Kecurangan Dengan Teknologi Informasi

Sebagai Variabel Moderasi. Jurnal Analisis, 7 (1), 91-98.

Kronologi Kasus Laporan Keuangan Garuda Indonesia hingga Kena Sanksi. (2019, Juni 28),

https://economy.okezone.com

Meenatkshi, R. and K. Sivaranjani. 2016. A Comparative Study on Fraud Detection in Financial

Statement using Data Mining Technique. International Journal of Computer Science and

Mobile Computing, 5 (7), 382-386.

Muliani, Christine. 2018. Pengaruh Karakteristik Individu, Jenjang Jabatan, Pengalaman Audit,

Tekanan Anggaran Waktu Terhadap Kemampuan Auditor Mendeteksi Fraud. Skripsi.

Semarang: Universitas Katolik Soegijapranata.

Ningtyas, Indri. et. al. 2018. Pengaruh Pengalaman, Keahlian, Dan Skeptisisme Profesional

Terhadap Pendeteksian Kecurangan (Studi Empiris Pada BPK RI Perwakilan Sumatera

Selatan). Akuntabilitas : Jurnal Penelitian dan Pengembangan Akuntansi, 12 (2), 113-124.

Novita, Ulfa. 2015. Pengaruh Pengalaman, Beban Kerja, dan Pelatihan Terhadap Skeptisme dan

Kemampuan Auditor dalam Mendeteksi Kecurangan (Studi Empiris Pada Auditor di BPKP

Perwakilan Provinsi Riau). Journal Faculty of Economics Riau University. 2 (1), 1-16.

Pengalaman Kerja Auditor. (2011, Agustus 16), https://zetzu.blogspot.com

Putra, G. S. A and A. A. N. B. Dwirandra. 2019. The Effect of Auditor Experience, Type of

Personality and Fraud Auditing Training on Auditors Ability in Fraud Detecting with

Professional Skepticism as a Mediation Variable. International Research Journal of

Management, IT & Social Sciences, 6 (2), 31-43.

Prasetia, Dwi. 2019. Keahlian Auditor, Independensi Dan Pengalaman Kerja Terhadap

Pendeteksian Kecurangan (Studi pada Badan Pengawasan Keuangan dan Pembangunan

(BPKP) Kota Semarang). Skripsi. Surakarta: Institut Agama Islam Negeri.

Sanjaya, Aviani. 2017. Pengaruh Skeptisisme Profesional, Independensi, Kompetensi, Pelatihan

Auditor, dan Resiko Audit Terhadap Tanggung Jawab Auditor Dalam Mendeteksi

Kecurangan. Jurnal Akuntansi Bisnis, 15 (1), 41-55 .

Sanger, Christin Lisa et. al. 2016. Pengaruh Pengalaman Audit, Keahlian Audit dan Tekanan

Ketaatan Terhadap Audit Judgment BPK RI Perwakilan Provinsi Sulawesi Utara. Jurnal

Accountability : Universitas Sam Ratulangi. 5 (2), 11-22.

Sari, Ellen Mutiara. 2017. Pengaruh Pengalaman dan Pelatihan Terhadap Kekeliruan Auditor

(Studi pada KAP Kota Bandung). Skripsi. Institutional Repositories & Scientific Journals.

Bandung: Perpustakaan Fakultas Ekonomi dan Bisnis Universitas Pasundan.

Sihombing, Evenri. et. al. 2019. The Effect of Forensic Accounting, Training, Experience, Work

Load And Professional Skeptic On Auditors Ability To Detect Of Fraud. International

Journal of Scientific & Technology Research, 8 (8), 474-480.

Standar Akuntansi Keuangan. 2020. PSAK 1 tentang Penyajian Laporan Keuangan. IAI

(https://iaiglobal.or.id).

Standar Profesi Akuntansi Publik. 2014. SA No.240 tentang Tanggung Jawab Auditor Terkait

dengan Kecurangan dalam Suatu Audit atas Laporan Keuangan. IAPI (https://iapi.or.id).

Sugiyono. 2018. Metode Penelitian Bisnis (Pendekatan Kuantiatif, Kualitatif, Kombinasi, dan

R&D). Bandung: CV. Alfabeta.

https://economy.okezone.com/

https://zetzu.blogspot.com/

https://iaiglobal.or.id/

https://iapi.or.id/



Sujarweni, V. Wiratna. 2015. Metodologi Penelitian Bisnin dan Ekonomi. Yogyakarta: PT. Pustaka

Baru.

Wati, Nurlela Lela. 2018. Metodologi Penelitian Terapan (Aplikasi SPSS, EVIEWS, Smart PLS dan

AMOS). Bekasi Barat: CV. Pustaka Amri.

https://www.ojk.go.id

https://www.bpkp.go.id

https://www.kap-hsm.com

https://www.ojk.go.id/

https://www.bpkp.go.id/

https://www.kap-hsm.com/

the effect of work experience, training and auditor's

Documents